I read some paper about this subject [ Yen-Chun Peng Optical Material 39 2015; Luis A. Agapito PHYSICAL REVIEW X 5, 011006 (2015) ... ] then they show how to do and some values.
The valence configurations of the atoms were 4s2 3d10 for Zn, 2s2 2p4 for O.
The energy cutoff : 380eV
The change in energy: d-5 eV/atom
maximum force: 0.03 eV/anstrong
maximum stress: 0.05 GPa
maximum displacement tolerances: 0.001 anstrong
The energy convergence criterion: d-6 eV
The Ud value for Zn-3d : 10 eV
The Up value for O-2p orbital: 7 eV
This is my scf file. If someone know where I was wrong please correct me!!!
Many thanks !!!
Dear Mr. Alexander,
I tried using HSE function on my calculation.
But the file scf.out send me a message with error :
Error in routine read_namelists (17)
reading namelist system
I checked the spelling and found no wrong.
Could you give me an advice?
I am sorry to ask you like this, because I am a new one in QE.
Many thanks !
< this is my scf.in file >
Dear Nga,
Please study these papers before any calculations.
http://journals.aps.org/prb/abstract/10.1103/PhysRevB.82.205102
http://journals.aps.org/prl/abstract/10.1103/PhysRevLett.102.226401
http://journals.aps.org/prb/abstract/10.1103/PhysRevB.85.155109
Best Regards
Jaafar
Dear Dr. Alexander P. Litvinchuk and Dr. Jaafar Jalilian thank you for always help on RG.
I want to ask, in case of hybrid calculation it takes so much time. Is there any way to accelerate calculation. If we do geometry optimization with PBE then other calculation like bandgap we can do in single point calculation with hybrid functional. Then it will take less time.
My query is that, is this right way of doing calculation ? Will results be same in both cases when we are optimizing our calculation with hybrid functional and without hybrid functional but single point calculation with hybrid functional.
Shilendra
forces may not work with DFT+U, and since you are using forc_conv_thr = 1.0d-4, it might be a problem.
except for that, you can try pbesol for obtaining better values.
regards
M Chaitanya Varma
@Nga Do,I see the input file here attached. But i see the input problem in a system card. look you aim to do DFT+U calculation,but you are introducing hybrid functional HSE why? is that possible to do DFT+U and hybrid functional i.e two different types of exchange correlation function simultaneously in a single calculation. As to me it is better one of the functional.
- Badges
- Science topic